FlexAttention for Efficient High-Resolution Vision-Language Models

Current high-resolution vision-language models encode images as high-resolution image tokens and exhaustively take all these tokens to compute attention, which significantly increases the computational cost. To address this problem, we propose FlexAttention, a flexible attention mechanism for efficient high-resolution vision-language models. Specifically, a high-resolution image is encoded both as high-resolution tokens and low-resolution tokens, where only the low-resolution tokens and a few selected high-resolution tokens are utilized to calculate the attention map, which greatly shrinks the computational cost. The high-resolution tokens are selected via a high-resolution selection module which could retrieve tokens of relevant regions based on an input attention map. The selected high-resolution tokens are then concatenated to the low-resolution tokens and text tokens, and input to a hierarchical self-attention layer which produces an attention map that could be used for the next-step high-resolution token selection. The hierarchical self-attention process and high-resolution token selection process are performed iteratively for each attention layer. Experiments on multimodal benchmarks prove that our FlexAttention outperforms existing high-resolution VLMs (e.g., relatively ~9% in V* Bench, ~7% in TextVQA), while also significantly reducing the computational cost by nearly 40%.

CoNav: A Benchmark for Human-Centered Collaborative Navigation

Human-robot collaboration, in which the robot intelligently assists the human with the upcoming task, is an appealing objective. To achieve this goal, the agent needs to be equipped with a fundamental collaborative navigation ability, where the agent should reason human intention by observing human activities and then navigate to the human's intended destination in advance of the human. However, this vital ability has not been well studied in previous literature. To fill this gap, we propose a collaborative navigation (CoNav) benchmark. Our CoNav tackles the critical challenge of constructing a 3D navigation environment with realistic and diverse human activities. To achieve this, we design a novel LLM-based humanoid animation generation framework, which is conditioned on both text descriptions and environmental context. The generated humanoid trajectory obeys the environmental context and can be easily integrated into popular simulators. We empirically find that the existing navigation methods struggle in CoNav task since they neglect the perception of human intention. To solve this problem, we propose an intention-aware agent for reasoning both long-term and short-term human intention. The agent predicts navigation action based on the predicted intention and panoramic observation. The emergent agent behavior including observing humans, avoiding human collision, and navigation reveals the efficiency of the proposed datasets and agents.

MAGIC: Map-Guided Few-Shot Audio-Visual Acoustics Modeling

Few-shot audio-visual acoustics modeling seeks to synthesize the room impulse response in arbitrary locations with few-shot observations. To sufficiently exploit the provided few-shot data for accurate acoustic modeling, we present a *map-guided* framework by constructing acoustic-related visual semantic feature maps of the scenes. Visual features preserve semantic details related to sound and maps provide explicit structural regularities of sound propagation, which are valuable for modeling environment acoustics. We thus extract pixel-wise semantic features derived from observations and project them into a top-down map, namely the **observation semantic map**. This map contains the relative positional information among points and the semantic feature information associated with each point. Yet, limited information extracted by few-shot observations on the map is not sufficient for understanding and modeling the whole scene. We address the challenge by generating a **scene semantic map** via diffusing features and anticipating the observation semantic map. The scene semantic map then interacts with echo encoding by a transformer-based encoder-decoder to predict RIR for arbitrary speaker-listener query pairs. Extensive experiments on Matterport3D and Replica dataset verify the efficacy of our framework.

3D-VLA: A 3D Vision-Language-Action Generative World Model

Recent vision-language-action (VLA) models rely on 2D inputs, lacking integration with the broader realm of the 3D physical world. Furthermore, they perform action prediction by learning a direct mapping from perception to action, neglecting the vast dynamics of the world and the relations between actions and dynamics. In contrast, human beings are endowed with world models that depict imagination about future scenarios to plan actions accordingly. To this end, we propose 3D-VLA by introducing a new family of embodied foundation models that seamlessly link 3D perception, reasoning, and action through a generative world model. Specifically, 3D-VLA is built on top of a 3D-based large language model (LLM), and a set of interaction tokens is introduced to engage with the embodied environment. Furthermore, to inject generation abilities into the model, we train a series of embodied diffusion models and align them into the LLM for predicting the goal images and point clouds. To train our 3D-VLA, we curate a large-scale 3D embodied instruction dataset by extracting vast 3D-related information from existing robotics datasets. Our experiments on held-in datasets demonstrate that 3D-VLA significantly improves the reasoning, multimodal generation, and planning capabilities in embodied environments, showcasing its potential in real-world applications.

MultiPLY: A Multisensory Object-Centric Embodied Large Language Model in 3D World

Human beings possess the capability to multiply a melange of multisensory cues while actively exploring and interacting with the 3D world. Current multi-modal large language models, however, passively absorb sensory data as inputs, lacking the capacity to actively interact with the objects in the 3D environment and dynamically collect their multisensory information. To usher in the study of this area, we propose MultiPLY, a multisensory embodied large language model that could incorporate multisensory interactive data, including visual, audio, tactile, and thermal information into large language models, thereby establishing the correlation among words, actions, and percepts. To this end, we first collect Multisensory Universe, a large-scale multisensory interaction dataset comprising 500k data by deploying an LLM-powered embodied agent to engage with the 3D environment. To perform instruction tuning with pre-trained LLM on such generated data, we first encode the 3D scene as abstracted object-centric representations and then introduce action tokens denoting that the embodied agent takes certain actions within the environment, as well as state tokens that represent the multisensory state observations of the agent at each time step. In the inference time, MultiPLY could generate action tokens, instructing the agent to take the action in the environment and obtain the next multisensory state observation. The observation is then appended back to the LLM via state tokens to generate subsequent text or action tokens. We demonstrate that MultiPLY outperforms baselines by a large margin through a diverse set of embodied tasks involving object retrieval, tool use, multisensory captioning, and task decomposition.

A Simple Knowledge Distillation Framework for Open-world Object Detection

Open World Object Detection (OWOD) is a novel computer vision task with a considerable challenge, bridging the gap between classic object detection (OD) benchmarks and real-world object detection. In addition to detecting and classifying seen/known objects, OWOD algorithms are expected to localize all potential unseen/unknown objects and incrementally learn them. The large pre-trained vision-language grounding models (VLM,eg, GLIP) have rich knowledge about the open world, but are limited by text prompts and cannot localize indescribable objects. However, there are many detection scenarios which pre-defined language descriptions are unavailable during inference. In this paper, we attempt to specialize the VLM model for OWOD task by distilling its open-world knowledge into a language-agnostic detector. Surprisingly, we observe that the combination of a simple knowledge distillation approach and the automatic pseudo-labeling mechanism in OWOD can achieve better performance for unknown object detection, even with a small amount of data. Unfortunately, knowledge distillation for unknown objects severely affects the learning of detectors with conventional structures for known objects, leading to catastrophic forgetting. To alleviate these problems, we propose the down-weight loss function for knowledge distillation from vision-language to single vision modality. Meanwhile, we decouple the learning of localization and recognition to reduce the impact of category interactions of known and unknown objects on the localization learning process. Comprehensive experiments performed on MS-COCO and PASCAL VOC demonstrate the effectiveness of our methods.

DCIR: Dynamic Consistency Intrinsic Reward for Multi-Agent Reinforcement Learning

Learning optimal behavior policy for each agent in multi-agent systems is an essential yet difficult problem. Despite fruitful progress in multi-agent reinforcement learning, the challenge of addressing the dynamics of whether two agents should exhibit consistent behaviors is still under-explored. In this paper, we propose a new approach that enables agents to learn whether their behaviors should be consistent with that of other agents by utilizing intrinsic rewards to learn the optimal policy for each agent. We begin by defining behavior consistency as the divergence in output actions between two agents when provided with the same observation. Subsequently, we introduce dynamic consistency intrinsic reward (DCIR) to stimulate agents to be aware of others' behaviors and determine whether to be consistent with them. Lastly, we devise a dynamic scale network (DSN) that provides learnable scale factors for the agent at every time step to dynamically ascertain whether to award consistent behavior and the magnitude of rewards. We evaluate DCIR in multiple environments including Multi-agent Particle, Google Research Football and StarCraft II Micromanagement, demonstrating its efficacy.

CoVLM: Composing Visual Entities and Relationships in Large Language Models Via Communicative Decoding

A remarkable ability of human beings resides in compositional reasoning, i.e., the capacity to make "infinite use of finite means". However, current large vision-language foundation models (VLMs) fall short of such compositional abilities due to their "bag-of-words" behaviors and inability to construct words that correctly represent visual entities and the relations among the entities. To this end, we propose CoVLM, which can guide the LLM to explicitly compose visual entities and relationships among the text and dynamically communicate with the vision encoder and detection network to achieve vision-language communicative decoding. Specifically, we first devise a set of novel communication tokens for the LLM, for dynamic communication between the visual detection system and the language system. A communication token is generated by the LLM following a visual entity or a relation, to inform the detection network to propose regions that are relevant to the sentence generated so far. The proposed regions-of-interests (ROIs) are then fed back into the LLM for better language generation contingent on the relevant regions. The LLM is thus able to compose the visual entities and relationships through the communication tokens. The vision-to-language and language-to-vision communication are iteratively performed until the entire sentence is generated. Our framework seamlessly bridges the gap between visual perception and LLMs and outperforms previous VLMs by a large margin on compositional reasoning benchmarks (e.g., ~20% in HICO-DET mAP, ~14% in Cola top-1 accuracy, and ~3% on ARO top-1 accuracy). We also achieve state-of-the-art performances on traditional vision-language tasks such as referring expression comprehension and visual question answering.

FGPrompt: Fine-grained Goal Prompting for Image-goal Navigation

Learning to navigate to an image-specified goal is an important but challenging task for autonomous systems. The agent is required to reason the goal location from where a picture is shot. Existing methods try to solve this problem by learning a navigation policy, which captures semantic features of the goal image and observation image independently and lastly fuses them for predicting a sequence of navigation actions. However, these methods suffer from two major limitations. 1) They may miss detailed information in the goal image, and thus fail to reason the goal location. 2) More critically, it is hard to focus on the goal-relevant regions in the observation image, because they attempt to understand observation without goal conditioning. In this paper, we aim to overcome these limitations by designing a Fine-grained Goal Prompting (FGPrompt) method for image-goal navigation. In particular, we leverage fine-grained and high-resolution feature maps in the goal image as prompts to perform conditioned embedding, which preserves detailed information in the goal image and guides the observation encoder to pay attention to goal-relevant regions. Compared with existing methods on the image-goal navigation benchmark, our method brings significant performance improvement on 3 benchmark datasets (i.e., Gibson, MP3D, and HM3D). Especially on Gibson, we surpass the state-of-the-art success rate by 8% with only 1/50 model size. Project page: https://xinyusun.github.io/fgprompt-pages

$A^2$Nav: Action-Aware Zero-Shot Robot Navigation by Exploiting Vision-and-Language Ability of Foundation Models

We study the task of zero-shot vision-and-language navigation (ZS-VLN), a practical yet challenging problem in which an agent learns to navigate following a path described by language instructions without requiring any path-instruction annotation data. Normally, the instructions have complex grammatical structures and often contain various action descriptions (e.g., "proceed beyond", "depart from"). How to correctly understand and execute these action demands is a critical problem, and the absence of annotated data makes it even more challenging. Note that a well-educated human being can easily understand path instructions without the need for any special training. In this paper, we propose an action-aware zero-shot VLN method ($A^2$Nav) by exploiting the vision-and-language ability of foundation models. Specifically, the proposed method consists of an instruction parser and an action-aware navigation policy. The instruction parser utilizes the advanced reasoning ability of large language models (e.g., GPT-3) to decompose complex navigation instructions into a sequence of action-specific object navigation sub-tasks. Each sub-task requires the agent to localize the object and navigate to a specific goal position according to the associated action demand. To accomplish these sub-tasks, an action-aware navigation policy is learned from freely collected action-specific datasets that reveal distinct characteristics of each action demand. We use the learned navigation policy for executing sub-tasks sequentially to follow the navigation instruction. Extensive experiments show $A^2$Nav achieves promising ZS-VLN performance and even surpasses the supervised learning methods on R2R-Habitat and RxR-Habitat datasets.